Phase III represents the definitive randomized controlled trial that compares a fully developed intervention with an appropriate alternative to show its efficacy. In other words, it is focused on evaluating the effects of the intervention on outcomes for individuals and a limited set of symptoms including psychosocial, biobehavioral, or clinical. This phase starts with the assumption that there is a well-crafted intervention based upon the preliminary evidence of benefit and feasibility garnered from Phases I and II.
In an efficacy trial, the primary concern is with enhancing the internal validity of the study design so that observed benefits can be attributed to the intervention rather than other potential confounding variables (e.g., access to or utilization of other services, spontaneous improvement). So, for example, samples tend to be homogeneous, interventionists tend to be “super” clinicians, and treatment exposure is tightly controlled. Thus, efficacy trials are designed to test interventions under controlled, ideal conditions. There are numerous evaluative designs that can be used in this study phase. As these are amply described in classic clinical trial texts, we highlight only the most common designs in Table 2.2 for ease of reference.